摘要
敏捷卫星机动能力的增强带来观测机会的增多和观测时间窗口的增长,具有很大的应用潜能。但管控中由于观测开始时间是一个具有连续值域的变量,传统非敏捷卫星调度问题的组合优化建模方法不再适用。介绍了敏捷卫星的时间依赖、可控与不可控事件混合、资源过度受限等特性;给出了时间约束网络概念及敏捷卫星调度中的时间约束推理问题描述;分析了时间约束网络与距离图的等价性,借鉴Johnson算法对稀疏网络的特殊处理,采用重赋权技术,提出针对敏捷卫星调度的改进循环检测算法检查时间约束;建立了分支剪枝搜索算法,以约束规划方式获得时间一致解。最后通过实例验证所提方法的有效性,实验结果表明该方法能够较好地实现敏捷卫星调度的时间约束的一致性检验和搜索求解。
The agile earth observation satellite （AEOS） is a new type of satellite with high application po- tential for its added observation opportunities and flexible time window from enhanced maneuverability. Howev- er, as the start time of ohservations is a continuous variable, traditional combination optimization modeling methods for non-agile satellites are not usable for it. The characteristics of temporal dependence, the mix of controllable and uncontrollable events, and resource＇ s over-subscribed are introduced. The temporal constraint network concept and the temporal reasoning problem of AEOS are presented. The equivalence character of tem- poral constraint network and distance graph and the Johnson algorithm for sparse networks are analyzed. Then, an improved cycle checking algorithm for AEOS scheduling based on re-evaluating is proposed, and a branch- prune algorithm based on constraint programming way is used to get a consistent solution. Finally some experi- ments are presented to prove the validity of the method. The results show the efficiency and validity of the method in consistent checking and searching of temporal constraints in AEOS scheduling.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2013年第6期1206-1211,共6页
Systems Engineering and Electronics
基金
国家自然科学基金(61203180
70801062
70971131
71031007
71071156
71101013
71101150
71201171)
湖南省研究生创新项目(CX2011B023)
国防科学技术大学博士创新项目(B110504)资助课题
关键词
调度
时间推理
循环检测
敏捷卫星
时间依赖
简单时间网络
scheduling
temporal reasoning
cycle checking
agile earth observation satellite （AEOS）
temporal dependence
simple temporal network （STN）
